From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Petr Mladek <pmladek@suse.com>,
linux-mm@kvack.org,
"Luis Claudio R. Goncalves" <lgoncalv@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
Mel Gorman <mgorman@techsingularity.net>,
Michal Hocko <mhocko@suse.com>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH] mm/page_alloc: Use write_seqlock_irqsave() instead write_seqlock() + local_irq_save().
Date: Fri, 23 Jun 2023 11:45:44 +0200 [thread overview]
Message-ID: <20230623094544.RcLNdg6H@linutronix.de> (raw)
In-Reply-To: <c70f4eae-4a81-4faa-c70c-b88cfda120f4@I-love.SAKURA.ne.jp>
On 2023-06-23 00:43:00 [+0900], Tetsuo Handa wrote:
> On 2023/06/23 0:04, Petr Mladek wrote:
> > AFAIK, rt_spin_lock(lock) fulfills exactly the above requirements.
> > The owner could schedule. The waiter could schedule as well so that
> > they could be running on the same CPU. Also the current owner gets
> > higher priority when the is a waiter with the higher priority to avoid
> > the priority inversion.
>
> Excuse me, but that is about multiple threads trying to acquire the same lock, isn't it?
>
> Our case is that one thread which makes zonelist_update_seq.seqcount odd acquires
> zonelist_update_seq.lock but threads spinning at
> read_seqbegin(zonelist_update_seq.seqcount) from zonelist_iter_begin() do nothing but
> cpu_relax() busy loop. There is no way to teach that these threads need to give
> CPU to the thread which made zonelist_update_seq.seqcount odd...
For !RT there is no spinning because interrupts are disabled.
For RT there is no spinning because the reader blocks on lock owned by
writer.
> > The result would look like:
> >
> > in kernel/linux/printk.h:
> >
> > static inline void printk_deferred_enter(void)
> > {
> > if (!defined(CONFIG_PREEMPT_RT))
> > preempt_disable();
> > else
> > migrate_disable();
> >
> > __printk_safe_enter();
> > }
> >
> > in mm/page_alloc.c
> >
> > printk_deferred_enter();
> > write_seqlock_irqsafe(&zonelist_update_seq, flags);
>
> OK. But for stable,
>
> + if (defined(CONFIG_PREEMPT_RT))
> + migrate_disable();
> /*
> * Explicitly disable this CPU's interrupts before taking seqlock
> * to prevent any IRQ handler from calling into the page allocator
> * (e.g. GFP_ATOMIC) that could hit zonelist_iter_begin and livelock.
> */
> local_irq_save(flags);
> /*
> * Explicitly disable this CPU's synchronous printk() before taking
> * seqlock to prevent any printk() from trying to hold port->lock, for
> * tty_insert_flip_string_and_push_buffer() on other CPU might be
> * calling kmalloc(GFP_ATOMIC | __GFP_NOWARN) with port->lock held.
> */
> printk_deferred_enter();
> write_seqlock(&zonelist_update_seq);
>
> will be easier to apply.
I would prefer not to worry about stable-RT but about upstream and then
we backport what is needed into the stable-RT trees once we settled on
something. This does not affect !RT.
Sebastian
next prev parent reply other threads:[~2023-06-23 9:45 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-21 10:40 [PATCH] mm/page_alloc: Use write_seqlock_irqsave() instead write_seqlock() + local_irq_save() Sebastian Andrzej Siewior
2023-06-21 10:59 ` Michal Hocko
2023-06-21 11:16 ` Sebastian Andrzej Siewior
2023-06-21 11:49 ` Michal Hocko
2023-06-21 13:11 ` Sebastian Andrzej Siewior
2023-06-21 13:22 ` Michal Hocko
2023-06-21 13:25 ` Sebastian Andrzej Siewior
2023-06-21 11:14 ` David Hildenbrand
2023-06-21 11:33 ` Tetsuo Handa
2023-06-21 12:40 ` Petr Mladek
2023-06-21 13:08 ` Sebastian Andrzej Siewior
2023-06-21 13:06 ` Sebastian Andrzej Siewior
2023-06-21 13:32 ` Tetsuo Handa
2023-06-21 14:34 ` Sebastian Andrzej Siewior
2023-06-21 14:50 ` Tetsuo Handa
2023-06-21 23:24 ` Tetsuo Handa
2023-06-22 7:18 ` Michal Hocko
2023-06-22 10:58 ` Tetsuo Handa
2023-06-22 12:09 ` Michal Hocko
2023-06-22 13:36 ` Tetsuo Handa
2023-06-22 14:11 ` Petr Mladek
2023-06-22 14:28 ` Tetsuo Handa
2023-06-23 9:35 ` Sebastian Andrzej Siewior
2023-06-22 15:04 ` Petr Mladek
2023-06-22 15:43 ` Tetsuo Handa
2023-06-23 9:45 ` Sebastian Andrzej Siewior [this message]
2023-06-23 9:51 ` Tetsuo Handa
2023-06-23 10:11 ` Sebastian Andrzej Siewior
2023-06-23 10:36 ` Tetsuo Handa
2023-06-23 12:44 ` Sebastian Andrzej Siewior
2023-06-23 12:57 ` Michal Hocko
2023-06-23 10:53 ` Petr Mladek
2023-06-23 11:16 ` Tetsuo Handa
2023-06-23 13:31 ` Sebastian Andrzej Siewior
2023-06-23 15:38 ` Petr Mladek
2023-06-23 16:04 ` Sebastian Andrzej Siewior
2023-06-23 9:31 ` Sebastian Andrzej Siewior
2023-06-23 7:27 ` Sebastian Andrzej Siewior
2023-06-21 15:38 ` Petr Mladek
2023-06-23 8:12 ` Sebastian Andrzej Siewior
2023-06-23 9:21 ` Michal Hocko
2023-06-23 9:58 ` Sebastian Andrzej Siewior
2023-06-23 10:43 ` Michal Hocko
2023-06-23 10:45 ` Sebastian Andrzej Siewior
2023-06-23 10:50 ` Sebastian Andrzej Siewior
2023-06-23 11:32 ` Michal Hocko
2023-06-23 10:40 ` Petr Mladek
2023-06-23 13:24 ` Sebastian Andrzej Siewior
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230623094544.RcLNdg6H@linutronix.de \
--to=bigeasy@linutronix.de \
--cc=akpm@linux-foundation.org \
--cc=lgoncalv@redhat.com \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@suse.com \
--cc=penguin-kernel@i-love.sakura.ne.jp \
--cc=pmladek@suse.com \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).